Columbia c/o 2015 Applicants (2011-2012 cycle) Forum

[KevinP]

In on Dec 5th! 3.65 and 164/171 LSAT (so yeah WHOA totally was not expecting it). I'm still half waiting for someone to email me and be like JUST KIDDING!!! but so far.....EEEEE!

Congratulations! Your post gives me hope.

[cogitoergosum]

I wouldn't be surprised to see Columbia dip down to 172/3.5 candidates just to maintain their median. (NYU already let go of their 172 median from a few years ago, so they're less constrained by that.)

So would it be reasonable to say that a 172 might be more statistically valuable to CLS than a 173?

[Themaddh]

I wouldn't be surprised to see Columbia dip down to 172/3.5 candidates just to maintain their median. (NYU already let go of their 172 median from a few years ago, so they're less constrained by that.)

So would it be reasonable to say that a 172 might be more statistically valuable to CLS than a 173?

How would that be the case?

[Boston_NYG2245]

Since LSATs are curved based on previous tests, would there necessarily be a a drop off in 170+ scores just because of the drop in people sitting for the exam? If the bulk of the people not taking are people who would have made sub 170 scores, there wouldn't be a change. That's working off of the assumption that the LSAT curve isn't actually a strict percentile breakdown of that test's set of takers, and it's probably still incredibly unlikely, just not technically impossible. I think. I'm bad at math; that's why I'm going to law school.

I'm not sure; what constitutes a certain score, especially on the tail end of the distribution (ie above 170) has never really changed much if you look at the numbers. I realize that it's hardly a statistically valid sample, but I can tell you that based on the number of questions I got wrong/right, my LSAT (taken in October) score fell right where you would expect it to. Projecting that out would imply that less people taking the test would yield less 170+ scores (as an absolute number not as a % of test takers). Am I making sense? Or totally off base...

[KevinP]

I think that the almost top-ranked schools will take longer to issue decisions this time around. The sharp drop off in the number of June and October LSAT test takers (16%+ so far) has resulted in a worrying contraction of the number of 170+ scores, which we can refer to as generally "admissible" scores for this caliber of law school. (Recall that LSAT scores are curved to reflect a certain percentile rank on the exam, and as such when the total number of test takers contracts, the number of high scores does as well.)

Yale and Harvard will not be materially affected by this contraction, as whatever admissible scores remain will matriculate at those schools, as they always do. Columbia, NYU, and Chicago, however, will have to fight even more vigorously than usual for the rest of the candidates to fill their classes. I suspect that admissions committees are waiting for December figures to come in (even an understanding of how many people took the test, to help hazard a guess as to how many new high scores are going to be minted) before they can issue more decisions; if it seems that a healthy number of new "admissible" candidates is waiting from the December administration, then there will be a greater number of deferrals. But regardless, I wouldn't be surprised to see Columbia dip down to 172/3.5 candidates just to maintain their median. (NYU already let go of their 172 median from a few years ago, so they're less constrained by that.)

With the outlook unclear, I suspect a larger number of ED candidates will be deferred and then accepted in April this cycle relative to previous ones. Still not a large number, but more than in the past. A more in-depth analysis of this phenomenon can be found in another thread, but I've thought about it for a while and I definitely think it explains the slower pace of decisions thus far.

I'm just hoping for a precipitous decline in the number of December LSAT administrations. =]

Interesting analysis. A couple things to note: the LSAT isn't curved, it's equated. Therefore, there exists a possibility that the drop in test takers was disproportionately from the lower end. Another interesting thing to note: The June/October test taker pool historically has always had the highest proportion of high scorers.

For example, here's data from 2009-2010:
June (Mean = 151.68, SD = 10.51). Approximate result: A 170 is the 95.9th percentile, ~4.1% of June test takers score a 170+.
December (Mean = 150.11, SD = 9.92). Approximate result: A 170 is the 97.8th percentile, ~2.2% of December test takers score a 170+.

However, I'm more inclined to think CLS would prefer a lock on their ED applicants instead of deferring/waitlisting them since CLS would lose the binding element. For example, I'm above CLS's median LSAT and slightly below CLS's median GPA. If CLS defers/waitlists me, I'll be a lot more likely to choose another T14 school if that school offers me money. I could be completely unrepresentative though.

[Tiago Splitter]

Since LSATs are curved based on previous tests, would there necessarily be a a drop off in 170+ scores just because of the drop in people sitting for the exam? If the bulk of the people not taking are people who would have made sub 170 scores, there wouldn't be a change. That's working off of the assumption that the LSAT curve isn't actually a strict percentile breakdown of that test's set of takers, and it's probably still incredibly unlikely, just not technically impossible. I think. I'm bad at math; that's why I'm going to law school.

Yeah the LSAT "curve" doesn't necessarily mean that the same proportion of people will score 170+. That said, the data from last year, when the number of test takers was down, shows that the drop happened pretty uniformly across all score ranges. Hopefully that trend continues.

[ahnhub]

Since LSATs are curved based on previous tests, would there necessarily be a a drop off in 170+ scores just because of the drop in people sitting for the exam? If the bulk of the people not taking are people who would have made sub 170 scores, there wouldn't be a change. That's working off of the assumption that the LSAT curve isn't actually a strict percentile breakdown of that test's set of takers, and it's probably still incredibly unlikely, just not technically impossible. I think. I'm bad at math; that's why I'm going to law school.

I'm not sure; what constitutes a certain score, especially on the tail end of the distribution (ie above 170) has never really changed much if you look at the numbers. I realize that it's hardly a statistically valid sample, but I can tell you that based on the number of questions I got wrong/right, my LSAT (taken in October) score fell right where you would expect it to. Projecting that out would imply that less people taking the test would yield less 170+ scores (as an absolute number not as a % of test takers). Am I making sense? Or totally off base...

The test is equated (it is designed to show how "smart" you are, not to pit you against everyone else sitting for the exam), so if a bunch of low scorers decided not to sit for the test, then yes, the absolute number of 170+s or whatever would stay the same. It never works out that way though. If 10,000 less people sit for the test, usually they fall all along the spectrum.

Also, from the historical info it would be pretty extraordinary for Columbia's median to fall below 172. That has been their median since 2007, I think. More likely the 25/75 take a hit, if anything.

[Boston_NYG2245]

Since LSATs are curved based on previous tests, would there necessarily be a a drop off in 170+ scores just because of the drop in people sitting for the exam? If the bulk of the people not taking are people who would have made sub 170 scores, there wouldn't be a change. That's working off of the assumption that the LSAT curve isn't actually a strict percentile breakdown of that test's set of takers, and it's probably still incredibly unlikely, just not technically impossible. I think. I'm bad at math; that's why I'm going to law school.

I'm not sure; what constitutes a certain score, especially on the tail end of the distribution (ie above 170) has never really changed much if you look at the numbers. I realize that it's hardly a statistically valid sample, but I can tell you that based on the number of questions I got wrong/right, my LSAT (taken in October) score fell right where you would expect it to. Projecting that out would imply that less people taking the test would yield less 170+ scores (as an absolute number not as a % of test takers). Am I making sense? Or totally off base...

The test is equated (it is designed to show how "smart" you are, not to pit you against everyone else sitting for the exam), so if a bunch of low scorers decided not to sit for the test, then yes, the absolute number of 170+s or whatever would stay the same. It never works out that way though. If 10,000 less people sit for the test, usually they fall all along the spectrum.

Thanks for the clarification.

[iamrobk]

Since LSATs are curved based on previous tests, would there necessarily be a a drop off in 170+ scores just because of the drop in people sitting for the exam? If the bulk of the people not taking are people who would have made sub 170 scores, there wouldn't be a change. That's working off of the assumption that the LSAT curve isn't actually a strict percentile breakdown of that test's set of takers, and it's probably still incredibly unlikely, just not technically impossible. I think. I'm bad at math; that's why I'm going to law school.

I'm not sure; what constitutes a certain score, especially on the tail end of the distribution (ie above 170) has never really changed much if you look at the numbers. I realize that it's hardly a statistically valid sample, but I can tell you that based on the number of questions I got wrong/right, my LSAT (taken in October) score fell right where you would expect it to. Projecting that out would imply that less people taking the test would yield less 170+ scores (as an absolute number not as a % of test takers). Am I making sense? Or totally off base...

The test is equated (it is designed to show how "smart" you are, not to pit you against everyone else sitting for the exam), so if a bunch of low scorers decided not to sit for the test, then yes, the absolute number of 170+s or whatever would stay the same. It never works out that way though. If 10,000 less people sit for the test, usually they fall all along the spectrum.

Also, from the historical info it would be pretty extraordinary for Columbia's median to fall below 172. That has been their median since 2007, I think. More likely the 25/75 take a hit, if anything.

Judging by what they, NYU, and a few other schools have done so far, I think they'd all rather let their GPA medians fall rather than their LSAT medians.

[AmoryB]

I think that the almost top-ranked schools will take longer to issue decisions this time around. The sharp drop off in the number of June and October LSAT test takers (16%+ so far) has resulted in a worrying contraction of the number of 170+ scores, which we can refer to as generally "admissible" scores for this caliber of law school. (Recall that LSAT scores are curved to reflect a certain percentile rank on the exam, and as such when the total number of test takers contracts, the number of high scores does as well.)

Yale and Harvard will not be materially affected by this contraction, as whatever admissible scores remain will matriculate at those schools, as they always do. Columbia, NYU, and Chicago, however, will have to fight even more vigorously than usual for the rest of the candidates to fill their classes. I suspect that admissions committees are waiting for December figures to come in (even an understanding of how many people took the test, to help hazard a guess as to how many new high scores are going to be minted) before they can issue more decisions; if it seems that a healthy number of new "admissible" candidates is waiting from the December administration, then there will be a greater number of deferrals. But regardless, I wouldn't be surprised to see Columbia dip down to 172/3.5 candidates just to maintain their median. (NYU already let go of their 172 median from a few years ago, so they're less constrained by that.)

With the outlook unclear, I suspect a larger number of ED candidates will be deferred and then accepted in April this cycle relative to previous ones. Still not a large number, but more than in the past. A more in-depth analysis of this phenomenon can be found in another thread, but I've thought about it for a while and I definitely think it explains the slower pace of decisions thus far.

I'm just hoping for a precipitous decline in the number of December LSAT administrations. =]

NYU's median dropped to 171? I was under the impression that it was still 172.

[addy11]

Dudes - I hate to be peremptory because I know a message board is about conversation, but all of this "How will the drop in LSAT takers affect me?" stuff has been explored in-depth in this thread (and others): http://top-law-schools.com/forums/viewt ... 6&t=170835

I direct you there because I know it's a fun/important thing to speculate about, and because of that it's very very easy to go very off topic.

[KevinP]

Judging by what they, NYU, and a few other schools have done so far, I think they'd all rather let their GPA medians fall rather than their LSAT medians.

As the number of test takers decreases, high LSAT scores will probably be more of a rarity than high GPAs. So, I can't say I'm too surprised.

NYU's median dropped to 171? I was under the impression that it was still 172.

Someone who attended NYU's orientation posted that the median dropped to 171, which is interesting considering class size decreased from 476 to 450. Just by looking at LSN, it seems like NYU was trying to protect that 172 median rather aggressively http://nyu.lawschoolnumbers.com/stats/1011/

[plurilingue]

.

[KevinP]

I knew there were a lot of neurotic posters lurking about! Great to see some activity on this board.

Hmm I was aware of the equating, but I thought the practical effect would be minimal due to the large number of test takers and general chance. I'm surprised to see it generate a difference of ~1.5% in a testing administration. Is this normal, or are you showing me a particularly divergent outcome? I know that June has stronger test-takers than October, and October than December, but I'm still a bit surprised... I'd love to see the data if you have it for other cycles.

Here's a link to the data:
http://www.lsac.org/LsacResources/Resea ... -10-03.pdf

Page 42 lists the means and standard deviations for multiple datasets, and page 44 has a nice visual representation for one of the years.

[Strange]

Hopefully this also means the "GPA floors" for places like Columbia drop a little and they show some more splitter love to maintain their LSAT median.

[cogitoergosum]

yep

[hyakku]

Hopefully this also means the "GPA floors" for places like Columbia drop a little and they show some more splitter love to maintain their LSAT median.

Fucking this. I'm praying to the gods this.

[TheRedMamba]

well I ED'd at 3.71/175 so it shows you what I know....

but Columbia was my first choice, way above chicago and NYU. I kinda figured my chances were slim at HYS so I wanted to lock up Columbia rather than risk falling back to chicago or nyu

[Take Two]

haha there are a lot worse things to be stuck with than a spot at CLS

[maxpower430]

Judging by what they, NYU, and a few other schools have done so far, I think they'd all rather let their GPA medians fall rather than their LSAT medians.

As the number of test takers decreases, high LSAT scores will probably be more of a rarity than high GPAs. So, I can't say I'm too surprised.

NYU's median dropped to 171? I was under the impression that it was still 172.

Someone who attended NYU's orientation posted that the median dropped to 171, which is interesting considering class size decreased from 476 to 450. Just by looking at LSN, it seems like NYU was trying to protect that 172 median rather aggressively http://nyu.lawschoolnumbers.com/stats/1011/

In one of the threads on the drop in test takers, I remember reading that the NYU administration said that they over-enrolled a year or two ago (I don't remember exactly), and that last year was an attempt to correct the imbalance. That said, they could have been trying to protect that median at the same time, but that was their reasoning for the drop in class size.

It will certainly be interesting to see how the cycle plays out everywhere, since as everyone has noted, the schools seem to be favoring LSAT over gpa (which makes sense given the subjective nature of GPAs). so it'd make a lot of sense for schools to be friendlier to splitters, and also less picky when it comes to softs. but only time will tell, good luck to everyone still waiting and congrats to all the acceptances

[AntipodeanPhil]

In one of the threads on the drop in test takers, I remember reading that the NYU administration said that they over-enrolled a year or two ago (I don't remember exactly), and that last year was an attempt to correct the imbalance. That said, they could have been trying to protect that median at the same time, but that was their reasoning for the drop in class size.

As far as I can tell, almost all of the top schools have lowered enrollment to some degree this year, and the few that have offered an explanation all seem to be saying it was just a correction for over-enrollment.

So, what's more plausible:

1. That they all coincidentally decided to correct for over-enrollment this year (often over-enrollment stretching back a few cycles), and it is just a coincidence they all did this in a year that applicant numbers dropped.

Or:

2. That no school wants to look bad by admitting its pool of applicants got worse.

[kwais]

CLS overenrolled this year, so I imagine that they will be a little cautious. They wanted to cut the class and got a huge yield. Not sure how or if that affects the LSAT discussion, but I thought I'd throw it in here.

[AntipodeanPhil]

CLS overenrolled this year, so I imagine that they will be a little cautious. They wanted to cut the class and got a huge yield. Not sure how or if that affects the LSAT discussion, but I thought I'd throw it in here.

Curious - I just checked the numbers, and it looks as if they admitted 9 more students last year than the year before. That's unusual. It makes me worry that Columbia might be willing to see the class size drop significantly this year.

[kwais]

CLS overenrolled this year, so I imagine that they will be a little cautious. They wanted to cut the class and got a huge yield. Not sure how or if that affects the LSAT discussion, but I thought I'd throw it in here.

Curious - I just checked the numbers, and it looks as if they admitted 9 more students last year than the year before. That's unusual. It makes me worry that Columbia might be willing to see the class size drop significantly this year.

yeah, they wanted 380 (I think) and ended up a good deal above that. they underestimated their brand I guess. i don't think it will negatively affect you guys much, I bet they'll just be slower

[maxpower430]

In one of the threads on the drop in test takers, I remember reading that the NYU administration said that they over-enrolled a year or two ago (I don't remember exactly), and that last year was an attempt to correct the imbalance. That said, they could have been trying to protect that median at the same time, but that was their reasoning for the drop in class size.

As far as I can tell, almost all of the top schools have lowered enrollment to some degree this year, and the few that have offered an explanation all seem to be saying it was just a correction for over-enrollment.

So, what's more plausible:

1. That they all coincidentally decided to correct for over-enrollment this year (often over-enrollment stretching back a few cycles), and it is just a coincidence they all did this in a year that applicant numbers dropped.

Or:

2. That no school wants to look bad by admitting its pool of applicants got worse.

Well, in NYU's case I think it's a bit murky, since we don't know if they could have maintained their medians at 476 matriculants, not to mention filling a class is a inexact science and class sizes vary from year to year. That said, it's likely a combination of both factors. From what I recall, many schools were over-enrolled because of the huge uptick in applicants two years ago, and if they actually want to get back to traditional levels, they have the added bonus of protecting their medians.